Breaking! Node::attr($attr, **$eval=false**, $to_str=false)

Author: duzun

README.md

@@ -161,16 +161,18 @@ if ( $banners ) {
 
     // Iterate over the result
     foreach($banners as $pos => $a) {
-        $links[$pos] = $a->attr('href'); // get absolute URL from href property
+        // $a->href property is the resolved $a->attr('href') relative to the
+        // documents <base href=...>, if present, or $doc->baseURL.
+        $links[$pos] = $a->href; // get absolute URL from href property
         $titles[$pos] = trim($a->text()); // strip all HTML tags and leave just text
 
         // Filter the result
         if ( !$a->hasClass('logo') ) {
-            // $a->style property is the parsed $a->attr('style')
+            // $a->style property is the parsed $a->attr('style'), same as $a->attr('style', true)
             if ( strtolower($a->style['position']) == 'fixed' ) continue;
 
             $img = $a->find('img')[0]; // ArrayAccess
-            if ( $img ) $images[$pos] = $img->src; // short for $img->attr('src')
+            if ( $img ) $images[$pos] = $img->src; // short for $img->attr('src', true)
         }
     }
 
@@ -190,6 +192,13 @@ $charset = $doc->charset;
 
 // Get the size of the document ( strlen($html) )
 $size = $doc->size;
+
+// The URL at which the document was requested
+$requestUri = $doc->href;
+
+// <base href=...>, if present, or the origin + dir path part from $doc->href.
+// The .href and .src props are resolved using this value.
+$baseURL = $doc->baseURL;
 ```
 
 Note: In case the charset meta attribute has a wrong value or the internal conversion fails for any other reason, `hQuery` would ignore the error and continue processing with the original HTML, but would register an error message on `$doc->html_errors['convert_encoding']`.

composer.json

@@ -2,7 +2,7 @@
   "name": "duzun/hquery",
   "type": "library",
   "description": "An extremely fast web scraper that parses megabytes of HTML in a blink of an eye. No dependencies. PHP5+",
-  "version": "3.2.0",
+  "version": "3.3.0",
   "license": "MIT",
   "authors": [
     {

package-lock.json

@@ -1,7 +1,7 @@
 {
   "name": "hquery.php",
   "description": "An extremely fast web scraper that parses megabytes of HTML in a blink of an eye. No dependencies. PHP5+",
-  "version": "3.2.0",
+  "version": "3.3.0",
   "author": {
     "name": "Dumitru Uzun",
     "email": "contact@duzun.me",

package.json

@@ -29,7 +29,7 @@ class_exists('duzun\\hQuery\\HTML_Index', false) or require_once __DIR__ . DIREC
  *
  *  Copyright (C) 2014-2018 Dumitru Uzun
  *
- *  @version 3.2.0
+ *  @version 3.3.0
  *  @author Dumitru Uzun (DUzun.ME)
  *  @license MIT
  */

src/hQuery.php

@@ -1,7 +1,6 @@
 <?php
 namespace duzun\hQuery;
 
-use duzun\hQuery\Parser\HTML;
 // ------------------------------------------------------------------------
 class_exists('duzun\\hQuery\\Node', false) or require_once __DIR__ . DIRECTORY_SEPARATOR . 'Node.php';
 
@@ -79,10 +78,6 @@ public function __get($name)
 
         switch ($name) {
             case 'style':
-                $style = $this->attr('style');
-                if ( !$style ) return self::$_ar_;
-                return $this->_prop[$name] = HTML::parseCssStr($style);
-
             case 'id':
             case 'class':
             case 'alt':
@@ -97,7 +92,7 @@ public function __get($name)
             // case 'search':
             // case 'hash':
             default:
-                return $this->attr($name);
+                return $this->attr($name, true);
         }
     }
 

src/hQuery/Element.php

@@ -546,7 +546,8 @@ public function _info()
             $lev[$b] = count($st);
         }
         foreach ($nm as $b => &$n) {
-            $n = str_repeat(' -', $lev[$b]) . ' < ' . $n . ' ' . $this->get_attr_byId($b, null, true) . ' >';
+            $n = str_repeat(' -', $lev[$b]) .
+                ' < ' . $n . ' ' . $this->get_attr_byId($b, null, true, false) . ' >';
         }
         $nm           = implode("\n", $nm);
         $inf['struc'] = $nm;
@@ -745,7 +746,7 @@ protected function _index_all()
         // Read <base href="..." /> tag
         if (!empty($this->tag_idx['base'])) {
             foreach ($this->tag_idx['base'] as $b => $e) {
-                if ($a = $this->get_attr_byId($b, 'href', false)) {
+                if ($a = $this->get_attr_byId($b, 'href', false, false)) {
                     $this->baseURI($a);
                     break;
                 }

src/hQuery/HTML_Index.php

@@ -15,7 +15,7 @@
 abstract class Node implements \Iterator, \Countable
 {
     // ------------------------------------------------------------------------
-    const VERSION = '3.2.0';
+    const VERSION = '3.3.0';
     // ------------------------------------------------------------------------
     /**
      * Response details of last request
@@ -125,17 +125,27 @@ public function __destruct()
      * Get and attribute or all attributes of first element in the collection.
      *
      * @param  string       $attr   attribute name, or NULL to get all
+     * @param  boolean      $eval if true, evaluate the attribute as a property,
+     *                            e.g. resolve .href & .src using document's baseURL
+     *                            or parse the .style property as an assoc arrays.
      * @param  boolean      $to_str When $attr is NULL, if true, get the list of attributes as string
      * @return array|string If no $attr, return a list of attributes, or attribute's value otherwise.
      */
-    public function attr($attr = null, $to_str = false)
+    public function attr($attr = null, $eval = false, $to_str = false)
     {
         $k = key($this->ids);
         if (null === $k) {
             reset($this->ids);
             $k = key($this->ids);
         }
-        return isset($k) ? $this->doc()->get_attr_byId($k, $attr, $to_str) : null;
+        if (!isset($k)) return null;
+
+        $val = $this->doc()->get_attr_byId($k, $attr, $to_str, !$eval);
+        if($eval && $attr == 'style') {
+            if (!$val) return self::$_ar_;
+            return HTMLParser::parseCssStr($val);
+        }
+        return $val;
     }
 
     // ------------------------------------------------------------------------
@@ -1114,7 +1124,13 @@ public function __get($name)
             return $this->_prop[$name];
         }
 
-        return $this->attr($name);
+        $val = $this->attr($name, true);
+
+        if($val && $name == 'style') {
+            $this->_prop[$name] = $val;
+        }
+
+        return $val;
     }
 
     /**

src/hQuery/Node.php

@@ -82,7 +82,7 @@ class hQueryCore extends PHPUnit_BaseClass
         href="//not-my-site.com/next.html"
         style="Color:blue;padding: 1px 2pt 3em 0; background-image:url(/path/to/img.jpg?url=param&and=another&one);"
     >Not My Site</a>
-    <img id="outerImg" src="https://cdn.duzun.me/images/logo.png" />
+    <img id="outerImg" src="//cdn.duzun.me/images/logo.png" />
 
     <dl id="dict1">
       <dt>Coffee</dt>
@@ -127,6 +127,21 @@ class hQueryCore extends PHPUnit_BaseClass
 
     public static $badHTML2 = '<html><head><meta http-equiv="Content-Type" content="text/html; charset=uft-8" /></head><body><a>A</a></body></html>';
 
+    public static $baseTag1 = '<!doctype html>
+<html>
+<head>
+    <meta content="/logo.png" property="og:image"/>
+    <base href="/base/path.html?how=rewrite#hash" />
+    <link rel="shortcut icon" href="/favicon.ico" class=pjax />
+</head>
+<body class="test-class">
+    <a href="rel-path/index.html" id="rel_path">relative path</a>
+    <a href="/abs-path/index.html" id="rel_origin">relative origin</a>
+    <a href="//not-my-site.com/next.html" id="rel_schema">relative schema</a>
+    <img id="rel_img" src="/images/logo.png" />
+</body>
+</html>';
+
     // Before any test
     public static function mySetUpBeforeClass()
     {
@@ -258,7 +273,8 @@ public function test_find()
         $this->assertTrue($a instanceof Element);
         $this->assertEquals('a', $a->nodeName);
         $this->assertEquals('link: This is a link', trim($a->text));
-        $this->assertEquals('https://DUzun.Me/path', $a->attr('href'));
+        $this->assertEquals('https://DUzun.Me/path', $a->href);
+        $this->assertEquals('/path', $a->attr('href'));
         $this->assertEquals('div', $a->parent->nodeName);
         $this->assertEquals('test-div', $a->parent->attr('id'));
 
@@ -438,29 +454,48 @@ public function test_hQuery_Element_ArrayAccess($doc)
     /**
      * @depends test_hQuery_Element_ArrayAccess
      */
-    public function test_attr($doc)
+    public function test_attr_and_prop($doc)
     {
+        // Note: there is no baseURI for $doc at this point.
         $e = $doc->find('#img1');
+        $a = $doc->find('a.aa:last');
 
         // It's magic!
         $this->assertEquals($e->src, $e->attr('src'));
         $this->assertEquals($e->src1, $e->attr('src1'));
         $this->assertEquals($e->src2, $e->attr('src2'));
+        $this->assertEquals($a->href, $a->attr('href'));
 
         // Standard way of accessing attributes:
         $this->assertEquals('/path/to/img.png', $e->attr('src'));
         $this->assertEquals('other/img/here.jpg', $e->attr('src2'));
         $this->assertEquals('//example.com/full/path.gif', $e->attr('src3'));
+        $this->assertEquals('#test', $a->attr('href'));
+
+        // $doc was loaded from the file "data/attr.html" and has no baseURI associated.
+        // Set the baseURI from document location, so that .href and .src props
+        // would be resolved.
+        $doc->location(self::$baseUrl);
+
+        // Properties are evaluated semantically:
+        $this->assertEquals(self::$baseUrl . 'path/to/img.png', $e->src);
+        $this->assertEquals('other/img/here.jpg', $e->src2); // .src2 ain't special
+
+        $this->assertEquals(self::$baseUrl . '#test', $a->href);
+        $this->assertEquals('#test', $a->attr('href'));
+
 
         // Relative vs Absolute URL paths
 
         // a[href] relative URL
         $a = self::$inst->find('a:first');
         $this->assertEquals(self::$baseUrl . 'path', $a->href);
+        $this->assertEquals('/path', $a->attr('href'));
 
         // a[href] absolute URL
         $a = self::$inst->find('a#outerLink');
         $this->assertEquals('https://not-my-site.com/next.html', $a->href);
+        $this->assertEquals('//not-my-site.com/next.html', $a->attr('href'));
 
         // $a->style is the parsed $a->attr('style'):
         $this->assertNotEmpty($a->style);
@@ -474,10 +509,12 @@ public function test_attr($doc)
         // img[src] absolute URL
         $a = self::$inst->find('img#outerImg');
         $this->assertEquals('https://cdn.duzun.me/images/logo.png', $a->src);
+        $this->assertEquals('//cdn.duzun.me/images/logo.png', $a->attr('src'));
 
         // link[href] relative URL
         $a = self::$inst->find('link', array('rel' => 'shortcut icon'));
         $this->assertEquals(self::$baseUrl . 'favicon.ico', $a->href);
+        $this->assertEquals('/favicon.ico', $a->attr('href'));
 
         // meta[content] - not a URL
         $m = self::$inst->find('meta', array('property' => 'og:image'));
@@ -493,7 +530,7 @@ public function test_attr($doc)
 
     // -----------------------------------------------------
     /**
-     * @depends test_attr
+     * @depends test_attr_and_prop
      */
     public function test_prop_charset($doc)
     {
@@ -515,20 +552,61 @@ public function test_prop_baseURL()
     {
         $baseURL = self::$inst->baseURL;
         $this->assertEquals(self::$baseUrl, $baseURL);
+
+        $doc = hQueryTestSurrogate::fromHTML(self::$baseTag1, self::$baseUrl . 'index.html');
+        $baseURL = $doc->baseURL;
+        $this->assertEquals(self::$baseUrl . 'base/', $baseURL);
+
+        $a = $doc->find('a#rel_path');
+        $this->assertEquals('rel-path/index.html', $a->attr('href'));
+        $this->assertEquals(self::$baseUrl . 'base/rel-path/index.html', $a->href);
+
+        $a = $doc->find('a#rel_origin');
+        $this->assertEquals('/abs-path/index.html', $a->attr('href'));
+        $this->assertEquals(self::$baseUrl . 'abs-path/index.html', $a->href);
+
+        $a = $doc->find('a#rel_schema');
+        $this->assertEquals('//not-my-site.com/next.html', $a->attr('href'));
+        $this->assertEquals('https://not-my-site.com/next.html', $a->href);
+
+        $img = $doc->find('img#rel_img');
+        $this->assertEquals('/images/logo.png', $img->attr('src'));
+        $this->assertEquals(self::$baseUrl . 'images/logo.png', $img->src);
+
+        return $doc;
     }
 
     // -----------------------------------------------------
-    public function test_prop_baseURI()
+    /**
+     * Either the <base href=...> or the location()
+     *
+     * @depends test_prop_baseURL
+     */
+    public function test_prop_baseURI($doc)
     {
         $baseURI = self::$inst->baseURI;
         $this->assertEquals(self::$baseUrl . 'index.html', $baseURI);
+
+        $baseURI = $doc->baseURI;
+        $this->assertEquals(self::$baseUrl . 'base/path.html?how=rewrite#hash', $baseURI);
+
+        return $doc;
     }
 
     // -----------------------------------------------------
-    // Alias of baseURI
-    public function test_prop_href()
+    /**
+     * URI at which the doc was accessed/loaded.
+     *
+     * @depends test_prop_baseURI
+     */
+    public function test_prop_href($doc)
     {
         $href = self::$inst->href;
+        $location = self::$inst->location();
+        $this->assertEquals($location, $href);
+        $this->assertEquals(self::$baseUrl . 'index.html', $href);
+
+        $href = $doc->href;
         $this->assertEquals(self::$baseUrl . 'index.html', $href);
     }